The long-term objectives of the proposed research are to develop antiviral drugs active against human cytomegalovirus (CMV). CMV is an important human pathogen, especially in patients with AIDS, for which there is a critical need to develop safe, effective antiviral agents. A rationale of this application is that such drugs can be discovered by detailed molecular characterization of previously identified drug targets--the UL97 protein that phosphorylates ganciclovir and the CMV DNA polymerase (Pol). Project 1 emphasizes x-ray crystallographic and computational modeling studies of these drug targets to design and identify new lead compounds. The initial focus is the UL97 protein, which is a protein kinase homolog. Following preliminary modeling based on known protein kinase structure, the structure of UL97 and/or an active fragment will be solved free and in complex with appropriate small molecules. The structures will be used in molecular modeling studies to design, synthesize, and test new anti-CMV drugs in collaboration with Burroughs Wellcome Company. As Project 2 proceeds (see below) the same approaches will be extended to the CMV Pol. Project 2 emphasized molecular genetic and biochemical approaches to the CMV Pol and the best candidate for its accessory protein, ICP36. These proteins and/or their functional domains will be overexpressed to provide materials for studies in Project 1. To aid in the interpretation of those studies, analyses of these proteins in terms of proteolytic domains and in terms of their interactions with DNA and substrates will be conducted using both biochemical and genetic approaches. A particular focus is the interaction of CMV Pol and ICP36. The interacting regions of the proteins will be mapped and peptides and other compounds that inhibit the interaction and its functional consequences will be derived. The information obtained regarding Pol and ICP36 in this project and in Project 1 will be used as the basis of the design of lead compounds for synthesis and testing by collaborators at Burroughs Wellcome. Thus, both Projects 1 and 2 take rational approaches to the development of new anti-CMV drugs directed against well established CMV drug targets.